Faion Product Manager

Product Domain Skill

Communication: User's language. Docs: English.

Purpose

Orchestrates product planning activities from ideation to MVP/MLP definition. Combines MLP planning and project bootstrap capabilities.

Merged From

Agents

Workflows

Workflow 1: Project Bootstrap

Full pipeline from idea to first task.

IDEA → CONCEPT → TECH STACK → MVP SCOPE → CONFIRMATION → BACKLOG → CONSTITUTION → TASK_000

Output:

aidocs/sdd/{project}/
├── constitution.md
├── roadmap.md
└── features/
    ├── backlog/{NN}-{feature}/spec.md
    └── 00-setup/tasks/todo/TASK_000_project_setup.md

Phases:

1. Vision Brainstorm - Apply Five Whys to get to real need
2. Tech Stack Selection - Frontend, backend, database
3. MVP Definition - Max 3-5 features, cut 50% nice-to-haves
4. User Confirmation - Present summary, get approval
5. Backlog Creation - Create feature specs
6. Constitution - Write project constitution
7. TASK_000 - Setup task with project goals

Workflow 2: MLP Planning

Transform MVP to Most Lovable Product.

1. Analyze MVP scope (competitors)
2. Extract current state from specs
3. Find MLP gaps
4. Propose WOW features
5. Update specs with MLP reqs
6. Create implementation order

Output:

product_docs/
├── mvp-scope-analysis.md
├── mlp-analysis-report.md
└── mlp-implementation-order.md

Methodologies (18)

M-PRD-001: MVP Scoping

Problem: Teams build too much before validation.

Framework:

1. Define core problem (one sentence)
2. Identify minimum features to solve it
3. Cut 50% of nice-to-haves
4. Set 4-week development constraint

Agent: faion-mvp-scope-analyzer-agent

M-PRD-002: MLP Planning

Problem: MVP works but users don't love it.

Framework:

1. Analyze MVP for friction points
2. Identify WOW moments opportunities
3. Plan delight features
4. Create implementation phases

MLP Dimensions:

• Delight: Micro-interactions, animations, polish
• Ease: Intuitive UX, zero friction
• Speed: Instant feedback, fast performance
• Trust: Security signals, reliability
• Personality: Brand voice, memorable moments

Agent: faion-mlp-agent (mode: find-gaps)

M-PRD-003: Feature Prioritization (RICE)

Problem: Too many features, limited resources.

Framework:

Formula: RICE = (Reach * Impact * Confidence) / Effort

Agent: faion-pm-agent

M-PRD-004: Feature Prioritization (MoSCoW)

Problem: Stakeholders disagree on priorities.

Framework:

Agent: faion-pm-agent

M-PRD-005: Roadmap Design

Problem: No clear development direction.

Framework:

1. Define time horizons (Now, Next, Later)
2. Align with business goals
3. Map features to milestones
4. Include buffer (20%)

Template:

## Q1 - Now (commit)
- Feature A: description
- Feature B: description

## Q2 - Next (plan)
- Feature C: description

## Q3+ - Later (vision)
- Theme: description

Agent: faion-pm-agent

M-PRD-006: User Story Mapping

Problem: Features lack user context.

Framework:

1. Define user activities (horizontal)
2. List tasks under activities
3. Prioritize vertically (top = MVP)
4. Draw release lines

Format:

Activity 1    Activity 2    Activity 3
---------     ---------     ---------
Task 1.1      Task 2.1      Task 3.1   ← MVP
Task 1.2      Task 2.2      Task 3.2   ← Release 2
Task 1.3                    Task 3.3   ← Release 3

Agent: faion-pm-agent

M-PRD-007: OKR Setting

Problem: Goals are vague or unmeasurable.

Framework:

Template:

## Objective: [What we want to achieve]

### KR1: [Metric] from X to Y
- Initiative: [Action]
- Initiative: [Action]

### KR2: [Metric] from X to Y
- Initiative: [Action]

Agent: faion-pm-agent

M-PRD-008: Problem Validation

Problem: Building solutions for non-problems.

Framework:

1. State the problem
2. Who has this problem?
3. How do they solve it today?
4. How much does it cost them?
5. Would they pay for a solution?

Evidence Required:

• 5+ user interviews
• Competitor analysis
• Search volume data

Agent: faion-market-researcher-agent

M-PRD-009: Assumption Mapping

Problem: Unknown risks in product decisions.

Framework:

               High Impact
                   │
    High Risk ─────┼───── Low Risk
    (test first)   │      (monitor)
                   │
               Low Impact

Steps:

1. List all assumptions
2. Rate impact (1-5)
3. Rate uncertainty (1-5)
4. Plot on matrix
5. Test high-impact, high-uncertainty first

Agent: faion-pm-agent

M-PRD-010: Lean Canvas

Problem: Business model not structured.

Framework:

┌─────────────┬──────────────┬─────────────┐
│ Problem     │ Solution     │ UVP         │
├─────────────┼──────────────┼─────────────┤
│ Key Metrics │ Unfair Adv.  │ Channels    │
├─────────────┴──────────────┴─────────────┤
│ Cost Structure      │ Revenue Streams    │
└─────────────────────┴────────────────────┘

Agent: faion-pm-agent

M-PRD-011: Jobs To Be Done

Problem: Features don't match user needs.

Framework:

When I [situation],
I want to [motivation],
So I can [expected outcome].

Example:

When I receive a payment notification,
I want to verify it's legitimate,
So I can confirm to my customer without worry.

Agent: faion-persona-builder-agent

M-PRD-012: Opportunity Scoring

Problem: Too many opportunities, no prioritization.

Framework:

Formula: Weighted average = final score

Agent: faion-market-researcher-agent

M-PRD-013: Value Proposition Canvas

Problem: Product-market fit unclear.

Framework:

Customer Segment:
├── Jobs (what they try to do)
├── Pains (obstacles, risks)
└── Gains (desired outcomes)

Value Map:
├── Products/Services
├── Pain Relievers (how we reduce pains)
└── Gain Creators (how we create gains)

FIT = Pain Relievers match Pains + Gain Creators match Gains

Agent: faion-market-researcher-agent

M-PRD-014: Sprint Planning

Problem: Unclear work for upcoming sprint.

Framework:

1. Review sprint goal
2. Select items from backlog
3. Break into tasks (< 8 hours each)
4. Estimate total capacity
5. Commit to sprint backlog

Velocity: Average story points from last 3 sprints

Agent: faion-pm-agent

M-PRD-015: Release Planning

Problem: No clear release timeline.

Framework:

1. Define release goal
2. Identify required features
3. Estimate total effort
4. Divide by velocity = sprints needed
5. Add 20% buffer
6. Set release date

Agent: faion-pm-agent

M-PRD-016: Backlog Refinement

Problem: Backlog items unclear or oversized.

Framework:

1. Review items for next 2-3 sprints
2. Add acceptance criteria
3. Estimate complexity
4. Split large items (> 13 points)
5. Prioritize

Definition of Ready:

• Clear description
• Acceptance criteria defined
• Estimated
• Dependencies identified
• Small enough for one sprint

Agent: faion-pm-agent

M-PRD-017: Five Whys Analysis

Problem: Root cause not understood.

Framework:

Problem: [State the problem]

Why 1: [First level answer]
Why 2: [Go deeper]
Why 3: [Go deeper]
Why 4: [Go deeper]
Why 5: [Root cause]

Action: [Fix the root cause]

Agent: faion-pm-agent

M-PRD-018: Impact Mapping

Problem: Features disconnected from goals.

Framework:

Goal → Actor → Impact → Deliverable
  │       │       │         │
  │       │       │         └─ What we build
  │       │       └─ How actor behavior changes
  │       └─ Who can help/hinder goal
  └─ Business objective

Example:

Goal: Increase revenue 20%
├─ Actor: Premium users
│  └─ Impact: Use more features
│     └─ Deliverable: Feature X
└─ Actor: Free users
   └─ Impact: Upgrade
      └─ Deliverable: Upgrade prompts

Agent: faion-pm-agent

Note: Full methodology details available in methodologies/ folder.

Execution

Bootstrap Flow

# Phase 1: Vision
AskUserQuestion(questions=[
    {"question": "Tell me about the project"},
])

# Phase 2: Tech Stack
AskUserQuestion([
    {"question": "Frontend?", "options": [
        {"label": "React + TypeScript"},
        {"label": "Next.js"},
        {"label": "None (API only)"}
    ]},
    {"question": "Backend?", "options": [
        {"label": "Python + Django"},
        {"label": "Python + FastAPI"},
        {"label": "Go"},
        {"label": "Node.js"}
    ]},
    {"question": "Database?", "options": [
        {"label": "PostgreSQL"},
        {"label": "SQLite"},
        {"label": "MongoDB"}
    ]}
])

# Phase 3: MVP Definition
Task(subagent_type="faion-mvp-scope-analyzer-agent",
     prompt=f"Analyze {product_type} competitors for MVP scope")

# Phase 4: Confirmation (MANDATORY)
# Present summary, get approval

# Phase 5: Create backlog and constitution

MLP Flow

# MVP analysis first
Task(subagent_type="faion-mvp-scope-analyzer-agent",
     prompt=f"Analyze {product_type} competitors for MVP scope")

# MLP orchestrator handles all phases
Task(subagent_type="faion-mlp-agent",
     prompt=f"""
     mode: analyze
     project_path: {project_path}
     """)

Task(subagent_type="faion-mlp-agent",
     prompt=f"""
     mode: find-gaps
     project_path: {project_path}
     """)

Task(subagent_type="faion-mlp-agent",
     prompt=f"""
     mode: propose
     project_path: {project_path}
     product_type: {product_type}
     """)

Task(subagent_type="faion-mlp-agent",
     prompt=f"""
     mode: update
     project_path: {project_path}
     """)

Task(subagent_type="faion-mlp-agent",
     prompt=f"""
     mode: plan
     project_path: {project_path}
     """)

Numbering Convention

Related Skills

Domain Skill v1.1 - Product Planning 18 Methodologies | 2 Agents Merged from: faion-mlp-planning, faion-project-bootstrap